Research ArticleIMMUNOLOGY

Altered 3D chromatin structure permits inversional recombination at the IgH locus

See allHide authors and affiliations

Science Advances  14 Aug 2020:
Vol. 6, no. 33, eaaz8850
DOI: 10.1126/sciadv.aaz8850
  • Fig. 1 Chromatin accessibility and transcription on WT and IGCR1-mutated IgH alleles.

    (A) Schematic map of IgH locus. Regulatory sequences are shown as colored ovals. Gene segments are indicated as colored boxes. Black lines under schematic refer to amplicons used in (D) to (G). (B) Capture Hi-C of WT (left) and IGCR1-deleted (middle) IgH alleles. Interacting regions are highlighted within dashed lines. Difference interaction map between WT and IGCR1-deleted IgH alleles is shown in the right. Decrease (blue) or increase (red) on IGCR1-deleted alleles is indicated. Position and orientation of CTCF-bound sites are indicated below heatmap (47). See also fig. S1A. (C) ATAC-seq assays of WT and IGCR1-mutated IgH alleles are shown (chr12: 114,554,576 to 114,839,712, mm9). Colored rectangles mark ATAC peaks that are (i) reduced by IGCR1 mutation (red), (ii) increased by IGCR1 mutation (green), or (iii) unaffected by IGCR1 mutation (black). Differential chromatin accessibility was quantified on the basis of moderated t tests using R package limma [*adjusted P value (false discovery rate) < 0.01]. Genomic localization and statistics of peaks are provided in fig. S1C. (D to G) RNA analyses of WT and IGCR1-mutated IgH alleles. Data are shown as means ± SEM of two (D, F, and G) or three (E) independent experiments.

  • Fig. 2 Recombination features of DST4.2 and DSP2 gene segments on WT and IGCR1-mutated IgH alleles.

    (A) IgH locus schematic showing location and orientation of primers used in the recombination assay. Rearrangements were assayed in bone marrow pro–B cells (B220+IgMCD43+) purified from WT and IGCR1-deficient mice (left) and in pro–B cell lines (right). ROSA26 served as the loading control. Data shown are representative of two independent experiments. (B) Recombination efficiency of 5′- and 3′-DH RSSs. Line 1 shows the organization of RSSs. 12- and 23-RSSs are shown as yellow and blue triangles, respectively. Recombination reporters contained an inverted EGFP gene flanked by a constant 23-RSS (from Jκ1) and test 12-RSSs from different DH gene segments (line 2). RAG1/2-induced recombination (line 3) permits EGFP expression (line 4). (C) Bar plots of the recombination efficiency of 5′- and 3′-DH RSSs. Controls include GFP expression from a reporter that lacks a functional 12-RSS (control 1) or in the absence of cotransfected RAG1/2 (control 2). (D) Ratio of 5′- or 3′-RSS utilization of indicated DH gene segments in 293T cotransfection assays. (E) Recombination efficiency assay in a RAG1/2-expressing pre–B cell line. EGFP, enhanced green fluorescent protein.

  • Fig. 3 Inversional recombination of VH81X to germline DH gene segments.

    (A) Schematic representation of VH81X rearrangements to 5′- and 3′-RSS of DQ52 gene segment by deletion (orange arrow) or inversion (black arrow). Locations and orientation of primers used to assay recombination are indicated. (B) Recombination assays of DQ52 by deletion or inversion from pro–B cell lines expressing RAG2 (top) or from bone marrow pro–B cells (bottom). Fivefold increasing amounts of genomic DNA starting at 8 ng (from cell lines) and 4 ng (from primary pro–B cells) were used as templates. ROSA26 served as the loading control. Data shown are representative of two biological replicate experiments. (C) Signal-end junctions were assayed by PCR as described for (B) using primers R1 and R2. (D to F) VH81X rearrangements to DSP2 gene segments (D) were assayed for inversional or deletional mechanisms (E) and signal-end junctions (F) as described for (A) to (C). ROSA26 served as the loading control. Data shown are representative of two biological replicate experiments.

  • Fig. 4 Inversional and deletional recombination of 3′-RSS of DQ52 and DSP2 on WT and IGCR1-mutated IgH alleles.

    Schematic representation of 3′-RSS of DQ52 (A) and DSP2 (B) rearrangements to VH81X gene segment by inversion (black arrows) or to JHs gene segments by deletion (blue arrows), respectively (left). Products of each form of rearrangements are shown to the right. RAG2-deficient pro–B cell lines with WT or IGCR1-mutated IgH alleles [CBE−/−(1) and CBE−/−(2)] were infected with a Rag2-expressing lentivirus, followed by genomic DNA purification after 14 days of selection with puromycin. LAM-HTGTS experiments were carried out as previously described (33, 39) with baits (red arrows) located 50- to 100-bp upstream of DQ52 (A) or DSP2 (B). Restriction enzyme Sacl-HF (R3156S, NEB) and BseYI (R0635S, NEB) were used to remove germline DNA with DQ52 and DSP2 as bait, respectively. Total reads were aligned to detect recombination by deletion to JHs and by inversion to VH81X. The lower reads of 3′ DSP2 RSS utilization compared to DQ52 gene may be due to inefficient restriction of germline DSP2 fragments during library preparation. Average reads and percentages from two independent experiments are shown in red.

  • Fig. 5 5′ and 3′ 12-RSS utilization in VHQ52.2.4 or VH7183.4.6-DH recombination.

    (A) Schematic of the 3′ IgH locus CTCF-binding sites (24) and mutations produced by CRISPR-Cas9 in the context of the IGCR1 mutated cell line CBE−/−(1). (B and D) Rearrangements assays of VHQ52.2.4 (B) or VH7183.4.6 (D) to 5′- or 3′-RSS of DQ52 by deletion (orange arrows) or inversion (black arrows), respectively. Locations and orientation of primers used to assay recombination are indicated, together with the 23-RSS of VHQ52.2.4 (B) or VH7183.4.6 (D) (blue triangles) and 12-RSSs flanking DQ52 gene segments (green and red triangles). Each set of three lanes contains fivefold increasing amounts of genomic DNA starting at 8 ng (lanes 3, 6, 9, 12, 15, and 18). ROSA26 was used as the loading control. Data shown are representative of two biological replicate experiments. (C and E) Rearrangements assays of VHQ52.2.4 (B) or VH7183.4.6 (D) to 5′- or 3′-RSS of DSP2 by deletion or inversion, respectively. ROSA26 was used as the loading control. Data shown are representative of two biological replicate experiments.

  • Fig. 6 Distinct rules of engagement during VH and DH gene segment rearrangements.

    (Top left) Configuration of WT unrearranged [germline (gl)] IgH alleles extending from the 3′VH genes until Eμ. Gray boxes, JH segments; colored boxes, DH segments; beige boxes, VH segments; blue triangles, 23-RSSs; yellow triangles, 12-RSSs. Previously proposed interactions between regulatory sequences Eμ, IGCR1, and a promoter 5′ of DQ52 (PQ52) are indicated. Asterisks identify CTCF-binding sites associated with proximal VH genes. Light green curved arrow signifies the previously proposed RAG1/2 scanning domain (10). The RAG1/2-rich RC maps closely with PQ52-Eμ region. (Top right) Proposed configuration of DFL16.1JH2 recombined WT IgH alleles. Eμ-IGCR1 interactions remain intact and the size of the RAG1/2 scanning domain is reduced (green arrow). (Bottom left) Configuration of germline IGCR1-mutated IgH alleles showing Eμ looping to the VH81X-associated CTCF-binding site. DST4.2 and VH7183.1.1 are now located within enlarged RAG1/2 scanning domain (green arrow). (Bottom right) Configuration of doubly mutated IgH alleles that lack IGCR1 as well as the VH81X-associated CTCF-binding site (81X) in which Eμ loops to the next available CTCF site located near VHQ52.2.4. DST4.2, VH7183.1.1, and VH81X are located within the further enlarged RAG1/2 scanning domain (green arrow).

Supplementary Materials

  • Supplementary Materials

    Altered 3D chromatin structure permits inversional recombination at the IgH locus

    Xiang Qiu, Fei Ma, Mingming Zhao, Yaqiang Cao, Lillian Shipp, Angela Liu, Arun Dutta, Amit Singh, Fatima Zohra Braikia, Supriyo De, William H. Wood III, Kevin G. Becker, Weiqiang Zhou, Hongkai Ji, Keji Zhao, Michael L. Atchison, Ranjan Sen

    Download Supplement

    This PDF file includes:

    • Figs. S1 to S6
    • Tables S1 to S4

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article